CtrlK
BlogDocsLog inGet started
Tessl Logo

perplexity

Web search and research using Perplexity AI. Use when user says "search", "find", "look up", "ask", "research", or "what's the latest" for generic queries. NOT for library/framework docs (use Context7) or workspace questions.

84

1.62x
Quality

76%

Does it follow best practices?

Impact

99%

1.62x

Average score across 3 eval scenarios

SecuritybySnyk

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./dist/plugins/perplexity/skills/perplexity/SKILL.md

The canonical home for this skill is perplexity in softaworks/agent-toolkit

SKILL.md
Quality
Evals
Security

Quality

Content

67%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-structured, actionable tool-selection skill with executable code and a clear priority chain. Its main weakness is redundancy: trigger terms and NOT-for guidance are restated multiple times across sections.

Suggestions

Collapse the repeated trigger-term list and NOT-for guidance into a single canonical location (e.g., Quick Reference) and reference it rather than restating in the opener, Examples, and Key Points.

Trim the Examples section, which largely restates the Quick Reference's NOT-Perplexity routing, to reduce token cost.

Consider moving the detailed Tool Selection Chain and Examples into a separate reference file to keep SKILL.md a lean overview.

DimensionReasoningScore

Conciseness

Mostly efficient with executable code blocks, but the trigger-term list and the NOT-for guidance are repeated across the opening line, Quick Reference, Examples, and Key Points sections, adding noticeable redundancy that could be tightened.

3 / 5

Actionability

Provides copy-paste-ready executable code with real parameters for both search and ask (e.g. max_results: 3, max_tokens_per_page: 512) and explicit when-to-increase-limits guidance, with only minor gaps in example coverage.

4 / 5

Workflow Clarity

The Tool Selection Chain gives a clear 1-7 priority order and the Quick Reference decision tree disambiguates Search vs Ask vs Researcher, yielding a clear selection sequence with only minor ambiguity.

4 / 5

Progressive Disclosure

Content is self-contained with no external references needed and is organized into clear labeled sections; at ~90 lines it exceeds the under-50-lines simple-skill threshold, so it sits at 'good structure, minor organization gaps' rather than a 5.

4 / 5

Total

15

/

20

Passed

Description

86%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description with excellent trigger-term coverage and clear what/when guidance plus explicit de-confliction boundaries. The only weakness is that the stated capabilities ('search', 'research') are generic rather than concretely enumerated.

Suggestions

Enumerate concrete capabilities (e.g., 'returns URLs and search results', 'synthesizes conversational answers') instead of the generic 'Web search and research'.

DimensionReasoningScore

Specificity

Names the domain (Perplexity AI) and two actions ("Web search and research") but the actions are generic rather than comprehensive concrete capabilities, matching the anchor 'Names domain and 1-2 concrete actions, but not comprehensive'.

3 / 5

Completeness

Explicitly answers both what ("Web search and research using Perplexity AI") and when ("Use when user says...") with concrete trigger phrases, plus a NOT-for boundary clause.

5 / 5

Trigger Term Quality

Provides comprehensive natural trigger phrases users would actually say ("search", "find", "look up", "ask", "research", "what's the latest") including synonyms, matching the comprehensive-coverage anchor.

5 / 5

Distinctiveness Conflict Risk

The explicit NOT-for boundaries (Context7, workspace questions) reduce conflict risk, but the trigger terms 'search'/'find' are inherently broad and overlap with WebSearch and Context7, leaving minor overlap risk with closely related skills.

4 / 5

Total

17

/

20

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
softaworks/agent-toolkit
Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.